Peer Cache Filling

Abstract

The embodiments provide peer cache filling. The peer cache filling allocates a set of caching servers to distribute content in response to user requests with a limited first subset of the set of servers having access to retrieve the content from an origin and with a larger second subset of the set of servers retrieving the content from the first subset of servers without accessing the origin. The peer cache filling dynamically escalates and deesclataes the allocation of the caching servers to the first and second subsets as demand for the content rises and falls. Peer cache filling is implemented by modifying request headers to identify designated hot content, provide a request identifier hash result for identifying the ordering of servers, and provide a value for designating which servers in the ordering as primary server with access to the origin.

Description

BACKGROUND ART[0001]A content delivery network (CDN) is a distributed platform that delivers customer (i.e., content provider) content to different end users from different distribution points within a digital network. The CDN operates different points-of-present (PoPs) throughout the digital network which form the content distribution points. One or more CDN caching servers operate in each of the PoPs. The CDN caching servers cache and serve the customer content on behalf of the CDN customers. The CDN effectively fans out the customer content from the customer origin to a larger set of CDN caching servers. The larger set of CDN caching servers then redistributes the customer content to an even larger set of end users while optimizing the delivery by localizing the distribution points from which end users receive the content from the CDN. The customer origin has minimal distribution capacity relative to the collective distribution capacity of the CDN as provided by the numerous CDN ...

AI Technical Summary

Benefits of technology

The patent describes a method for reducing the load that a content delivery network (CDN) places on a customer origin server by dynamically allocating additional caching servers to handle hot content. The CDN typically designates one or more caching servers from a larger number of caching servers in each PoP to distribute content. However, when demand for hot content spikes, the CDN may allocate additional caching servers from different PoPs, which can overwhelming the customer origin server. The patent proposes a solution to this problem by enabling the dynamically allocated servers to efficiently pre-fetch or retrieve in real-time the hot content that originates from the customer origin server while simultaneously reducing or eliminating the load that the dynamically allocated servers impose on the customer origin server. This allows the CDN to dynamically scale to allocate additional resources to satisfy high demand for hot content while maintaining a shield protecting the customer origin from excess load.

Problems solved by technology

Nonetheless, the CDN imposes its own load on the customer origin servers.

The problem with this dynamic allocation of additional caching servers to handle hot content is that each newly allocated caching server imposes additional load on the customer origin server.

For sufficiently hot content, the demand from the dynamically allocated caching servers from different CDN PoPs can overwhelm the customer origin.

This problem is exacerbated when the customer content is regularly updated and can only be cached for a short period of time, or contains dynamic elements that require retrieval from the customer origin for each request, for each new session, or for each new user.

Images